MLZ02_plot_dprime

MLZ02_plot_dprime.m
Get started
Design parameters
Load the data
Quick-and-dirty plot of the individual-subject data
Median response times, individual data
Time- and condition-related variables -- See MLPilot02_params for details
Group dprime plot:
Plot group averaged d'
Plot group d' by easy / hard delta
Plot group d' for delta 11 (easy_dprime2)
Plot group d' for delta 8 (hard_dprime2)
Plot group d' for delta 11 (easy_dprime2)
Sbj Order Learning Effect Scatter plot
Group Order Learning Effect bar plot
Group Order Learning Effect bar plot
Main effect of Order on dprime:
Group d'prime error plot
Group criterion plot:
Group RT plot
Main effect of Order on RT
Adjusted group RT plot

MLZ02_plot_dprime.m

Plot d' and RT learning curves for MLSecond02 experiment

File: work/MLExper/MLSecond02/prog/MLZ02_plot_dprime.m
Usage: publish('MLZ02_plot_dprime.m','html') ;    % 'latex','doc','ppt'
Date: 2009-08-21

% 1.0.2 2009-08-21 TH: Error bars on the group d' plot
% 1.0.1 2009-07-25 AP: Minor tweaks
% 1.0.0 2009-02-06 TH: Initial import to svn://orion/experiments/MLExper/

Get started

cd(fullfile(MLSecond02_pathstr,'data')) ;
fprintf('\n\nMLZ02_plot_dprime executed on %s.\n\n',datestr(now)) ;
fprintf('cd %s\n',pwd) ;

clear all ;


MLZ02_plot_dprime executed on 01-Sep-2009 15:01:50.

cd /Users/apetrov/a/r/w/work/MLExper/MLSecond02/data

Design parameters

Load them from the code that administered the experiment:

P = MLZ02_params(0) ;
design_params = P.design_params ;

N_sessions = design_params.max_session ;          % 6 sessions
blocks_session = design_params.blocks_session ;   % 8 blocks per session
N_blocks = N_sessions * blocks_session ;
trials_block = design_params.trials_block ;       % 104 trials per block
N_trials = N_blocks * trials_block ;
group_descr = {'1stN','2ndN','1stP','2ndP'} ;

deltas = design_params.deltas ;      % [11 8] = [easy difficult]

Load the data

filename = fullfile(MLSecond02_pathstr,'data','H.mat') ;
%recalculatep = true ;
recalculatep = false ;
if (recalculatep || ~exist(filename,'file'))
    %- Write down repository revision number
    fprintf('!/usr/local/bin/svn info \n') ;
    !/usr/local/bin/svn info
    %- Concatenate the individual data files into one master ASCII file
    fprintf('!cat sbj*.dat > raw_data.dat \n') ;
    !cat sbj*.dat > raw_data.dat
    %- Import to Matlab
    fprintf('\nD=import_MLZ02_data ...') ;
    D=import_MLZ02_data(fullfile(MLSecond02_pathstr,'data','raw_data.dat')) ;
    %- Calculate d-primes
    fprintf('\nH = MLZ02_hitcount(D) ...\n') ;
    [H,P] = MLZ02_hitcount(D,1,P) ;
    save(filename,'H') ;
    fprintf('\nsave %s \n\n',filename) ;
else
    fprintf('load %s \n\n',filename) ;
    load(filename) ;
end

% Used to factor out anomalous sbjs data comment out if you want
% non-adjusted H values
% adj_H-  Adjusted H that modifies the following sbjs data
%         1.  Sbj 334 anomalous RT block during session 3 block 5-6
%             Adjustment = Subtract constant
%         2.  Sbj 316 anomalous RT all of session 6
%             Adjustment = Subtract 600, divide by 3, add 600
%         3.  Sbj 325 anomalous d prime all of session 3
%             Adjustment = Add constant

% Set adj_H equal to H
adj_H = H;

% Make adjustment for sbj 334 alter RT_descr(don't change STD)
% replace in adj_H
idx = adj_H(1,16).RT_descr(21:22,1:7)-200;
idx(1:2,2) = idx(1:2,2)+200;
adj_H(1,16).RT_descr(21:22,1:7) = idx;

% Make adjustment for sbj 316 alter RT_descr(don't change STD) and replace
% in adj_H
idx = ((adj_H(1,5).RT_descr(41:48,1) -600)./3)+600;
idx2 = adj_H(1,5).RT_descr(41:48,2);
idx3 = ((adj_H(1,5).RT_descr(41:48,3:7) -600)./3)+600;
idx4 = zeros(8,7);
idx4(1:8,1) = idx;
idx4(1:8,2) = idx2;
idx4(1:8,3:7) = idx3;
adj_H(1,5).RT_descr(41:48,1:7) = idx4;

% Make adjustment for sbj 325 alter all_dprime2 and replace in adj_H
idx = adj_H(1,11).all_dprime2(9:12,1)+.95;
adj_H(1,11).all_dprime2(9:12,1) = idx;

% Set H equal to adj_H
H = adj_H;

load /Users/apetrov/a/r/w/work/MLExper/MLSecond02/data/H.mat

Quick-and-dirty plot of the individual-subject data

N_sbj = length(H) ;
[r,c] = recommend_subplot_size(N_sbj) ;
M = 3.5 ;
for k=1:N_sbj
    subplot(r,c,k) ;
    plot(H(k).block2,H(k).all_dprime2,'b.-'); hold on;
%   plot(H(k).block2,H(k).easy_dprime2,'m--',...   %1:5 pure 6:19 mixed 20:25pure
%        H(k).block2,H(k).hard_dprime2,'g--') ;    %26:28 mixed

    axis([0 N_blocks 0 M]) ; grid on ;
    set(gca,'xtick',(0:blocks_session:N_blocks),'xticklabel',[],'ytick',(0:ceil(M))) ;
    gr1 = H(k).group ;
    title(sprintf('S%d, g%d %s',H(k).sbj,gr1,group_descr{gr1})) ;
end
gr = [H.group] ;
xtab1(gr')

   Value   Count  Percent  Cum_cnt  Cum_pct
-------------------------------------------
       1       5    25.00        5    25.00
       2       5    25.00       10    50.00
       3       5    25.00       15    75.00
       4       5    25.00       20   100.00
-------------------------------------------

Median response times, individual data

clf
M = 1300 ;  % msec
m = 300 ;
for k=1:N_sbj
    subplot(r,c,k) ;
    idx = (2:2:H(k).N) ;
    RT_descr = H(k).RT_descr ;
    medianRT2 = (RT_descr(idx-1,5)+RT_descr(idx,5))./2 ;
    plot(H(k).block2,medianRT2,'r.-') ;
    axis([0 N_blocks m M]) ; grid on ;
    %ylabel('median RT') ;
    set(gca,'xtick',(0:blocks_session:N_blocks),'xticklabel',[]) ;
    gr1 = H(k).group ;
    title(sprintf('S%d, g%d %s',H(k).sbj,gr1,group_descr{gr1})) ;
end

% Discard incomplete data sets and pack the dprime2 field into a matrix



N = [H(:).N] ;
full_H = H ;
H = H(N==N_blocks) ;
N_sbj = length(H) ;
[r,c] = recommend_subplot_size(N_sbj) ;

gr = [H.group] ;
xtab1(gr')

dpr2 = [H.all_dprime2] ;      % [N_blocks/2 x N_subjects]
crit2 = [H.all_crit2] ;
easy2= [H.easy_dprime2] ;
ecrit2= [H.easy_crit2] ;
hard2= [H.hard_dprime2] ;
hcrit2= [H.hard_crit2] ;

RT_descr=cat(3,H.RT_descr) ;               % [N_blocks x 7 x N_subjects]
medianRT = squeeze(RT_descr(:,5,:)) ;      % [N_blocks x N_subjects]
idx = 2:2:N_blocks ;
medianRT2 = (medianRT(idx-1,:)+medianRT(idx,:))./2 ;

   Value   Count  Percent  Cum_cnt  Cum_pct
-------------------------------------------
       1       5    25.00        5    25.00
       2       5    25.00       10    50.00
       3       5    25.00       15    75.00
       4       5    25.00       20   100.00
-------------------------------------------

Time- and condition-related variables -- See MLPilot02_params for details

bl = H(1).block2 ;     % [28x1], blocks aggregated two by two
globalT = (bl-.5) .* trials_block ;

Group dprime plot:

Ss = cell(4,1) ;
l = {'Gr 1: 1stN','Gr 2: 2ndN','Gr 3: 1stP','Gr 4: 2ndP'} ;
for k = 1:4
    Ss{k} = find(gr==k) ;
    mdpr2(:,k) = mean(dpr2(:,Ss{k}),2) ;    % average over subjects
    mcrit2(:,k) = mean(crit2(:,Ss{k}),2) ;
    mRT2(:,k) = mean(medianRT2(:,Ss{k}),2) ;
    measy2(:,k) = mean(easy2(:,Ss{k}),2) ;
    mhard2(:,k) = mean(hard2(:,Ss{k}),2) ;


end
assert(isempty(intersect(Ss{1},Ss{2}))) ;   % Sanity check
assert(isempty(intersect(Ss{2},Ss{3}))) ;   % Sanity check
assert(isempty(intersect(Ss{3},Ss{4}))) ;   % Sanity check

clf
M = 3.5 ;
plot(globalT,mdpr2(:,1),'rs-',...
     globalT,mdpr2(:,2),'bo-',...
     globalT,mdpr2(:,3),'r^-',...
     globalT,mdpr2(:,4),'b^-') ;
axis([0 N_trials 0 M]) ; grid on ;
set(findobj(gca,'Color','r'),'MarkerFaceColor','r') ;
set(gca,'xtick',(0:blocks_session:N_blocks)*trials_block,'ytick',(0:ceil(M))) ;
xlabel('Trial') ; ylabel('d''') ;
title(sprintf('%s, %d Ss; %s, %d Ss; %s, %d Ss; %s, %d Ss',...
      l{1},length(Ss{1}),l{2},length(Ss{2}),...
      l{3},length(Ss{3}),l{4},length(Ss{4}))) ;
legend(l,'location','NorthWest') ;

Plot group averaged d'

gr = [H.group] ;
order1 = H(5).order2 ;
order2 = H(1).order2 ;


% Create a single averaged 28 x 1 matrix for gr 1+3 gr 2+4 and gr 1+2+3+4
% Group Averages all_dprime2
g13 = (mdpr2(:,1) + mdpr2(:,3)) / 2 ;
g24 = (mdpr2(:,2) + mdpr2(:,4)) / 2 ;
%g1234 = (mdpr2(:,1) + mdpr2(:,2) + mdpr2(:,3) + mdpr2(:,4)) / 4 ;

% Group Averages easy_dprime2 and hard_dprime2
g13easy = (measy2(:,1) + measy2(:,3)) / 2 ;
g13hard = (mhard2(:,1) + mhard2(:,3)) / 2 ;
g24easy = (measy2(:,2) + measy2(:,4)) / 2 ;
g24hard = (mhard2(:,2) + mhard2(:,4)) / 2 ;

pre_test  = (1:4)' ;   % pre-test blocks
pre_tn = (3:4)' ;      % pre-test training order
pre_tf = (1:2)' ;      % pre-test transfer order
post_test = (17:20)' ; % post-test blocks
post_tn = (17:18)' ;   % post-test training order
post_tf = (19:20)' ;   % post-test transfer order
tn = (3:18)' ;         % train in respective order
tn_opp = (19:24)' ;    % train opposite order

globalT = (bl-.5)*trials_block ;

% Loop to find the mean d_prime for all subjects in block2
for k=1:N_sbj
x(k) = mean(H(k).all_dprime2);
end;

% Loop that subtracts the mean from each point in all_dprime2
% Sets all sbj mean dprime to zero
% 20 subjects x 24 adjusted dprime values
for k=1:20
U(k,:) = H(k).all_dprime2(:) - x(k);
end

% Separate into train first and train second groups / create index
gr = [H.group];
gr = gr';
idx1 = find(gr==1 | gr==3);
idx2 = find(gr==2 | gr==4);
U1 = U(idx1,:);
U2 = U(idx2,:);


% Loop to compute standard error for each block for group1
for k=1:24
E(k) = std(U1(:,k));
ER1(k) = E(k)/sqrt(10);
end

% Loop to compute standard error for each block for group2
for k=1:24
E(k) = std(U2(:,k));
ER2(k) = E(k)/sqrt(10);
end

JTR = 10 ;   % horizontal jitter [in trials] for visibility
plot(-1,-1,'ks',-1,-1,'ko',-1,1,'rs',-1,1,'bo',...   % for the legend
     globalT(pre_tn)-JTR,g13(pre_tn),'rs',...
     globalT(pre_tf)-JTR,g13(pre_tf),'ro',...
     globalT(pre_tn)+JTR,g24(pre_tn),'bs',...
     globalT(pre_tf)+JTR,g24(pre_tf),'bo',...
     globalT(tn)-JTR,g13(tn),'rs-',...
     globalT(tn)+JTR,g24(tn),'bs-',...
     globalT(post_tn)-JTR,g13(post_tn),'rs',...
     globalT(post_tf)-JTR,g13(post_tf),'ro',...
     globalT(post_tn)+JTR,g24(post_tn),'bs',...
     globalT(post_tf)+JTR,g24(post_tf),'bo',...
     globalT(tn_opp)-JTR,g13(tn_opp),'ro-',...
     globalT(tn_opp)+JTR,g24(tn_opp),'bo-') ;
hold on ; errorbar1(globalT-JTR,g13,ER1,'rn'); hold off
hold on ; errorbar1(globalT+JTR,g24,ER2,'bn'); hold off


M = 2.5 ;
axis([0 N_trials 0 M]) ; grid on ;
set(gca,'xtick',(0:blocks_session:N_blocks)*trials_block,'ytick',(0:ceil(M))) ;
set(findobj(gca,'Color','r'),'MarkerFaceColor','r');
%set(findobj(gca,'Color','g'),'MarkerFaceColor','g');
xlabel('Trial') ; ylabel('d''') ;
%title(sprintf('Group Average Pre/Post-Test, Train, and Train Opposite'));
legend('Training order','Transfer order','1st order groups','2nd order groups','Location','NorthWest') ;
title(sprintf('MLSecond02: Groups 1+3 (%d Ss) and 2+4 (%d Ss)',...
      length(Ss{1})+length(Ss{3}), length(Ss{2})+length(Ss{4}))) ;

Plot group d' by easy / hard delta

gr = [H.group] ;
order1 = H(5).order2 ;
order2 = H(1).order2 ;


% Create a single averaged 28 x 1 matrix for gr 1+3 gr 2+4 and gr 1+2+3+4
% Group Averages all_dprime2
g13 = (mdpr2(:,1) + mdpr2(:,3)) / 2 ;
g24 = (mdpr2(:,2) + mdpr2(:,4)) / 2 ;
%g1234 = (mdpr2(:,1) + mdpr2(:,2) + mdpr2(:,3) + mdpr2(:,4)) / 4 ;

% Group Averages easy_dprime2 and hard_dprime2
g13easy = (measy2(:,1) + measy2(:,3)) / 2 ;
g13hard = (mhard2(:,1) + mhard2(:,3)) / 2 ;
g24easy = (measy2(:,2) + measy2(:,4)) / 2 ;
g24hard = (mhard2(:,2) + mhard2(:,4)) / 2 ;

pre_test  = (1:4)' ;   % pre-test blocks
pre_tn = (3:4)' ;      % pre-test training order
pre_tf = (1:2)' ;      % pre-test transfer order
post_test = (17:20)' ; % post-test blocks
post_tn = (17:18)' ;   % post-test training order
post_tf = (19:20)' ;   % post-test transfer order
tn = (3:18)' ;         % train in respective order
tn_opp = (19:24)' ;    % train opposite order

globalT = (bl-.5)*trials_block ;



%%Plot the group-averaged d'
clf
M = 3.5 ;
plot(-1,-1,'ks',-1,-1,'ko',-1,-1,'k-',-1,-1,'k--',...
    -1,1,'rs',-1,1,'bo',...   % for the legend
     globalT(pre_tn),g13easy(pre_tn),'rs',...
     globalT(pre_tn),g13hard(pre_tn),'rs--',...
     globalT(pre_tf),g13easy(pre_tf),'ro',...
     globalT(pre_tf),g13hard(pre_tf),'ro--',...
     globalT(pre_tn),g24easy(pre_tn),'bs',...
     globalT(pre_tn),g24hard(pre_tn),'bs--',...
     globalT(pre_tf),g24easy(pre_tf),'bo',...
     globalT(pre_tf),g24hard(pre_tf),'bo--',...
     globalT(tn),g13easy(tn),'rs-',...
     globalT(tn),g13hard(tn),'rs--',...
     globalT(tn),g24easy(tn),'bs-',...
     globalT(tn),g24hard(tn),'bs--',...
     globalT(post_tn),g13easy(post_tn),'rs',...
     globalT(post_tn),g13hard(post_tn),'rs--',...
     globalT(post_tf),g13easy(post_tf),'ro',...
     globalT(post_tf),g13hard(post_tf),'ro--',...
     globalT(post_tn),g24easy(post_tn),'bs',...
     globalT(post_tn),g24hard(post_tn),'bs--',...
     globalT(post_tf),g24easy(post_tf),'bo',...
     globalT(post_tf),g24hard(post_tf),'bo--',...
     globalT(tn_opp),g13easy(tn_opp),'ro-',...
     globalT(tn_opp),g13hard(tn_opp),'ro--',...
     globalT(tn_opp),g24easy(tn_opp),'bo-',...
     globalT(tn_opp),g24hard(tn_opp),'bo--')

axis([0 N_trials 0 M]) ; grid on ;
set(gca,'xtick',(0:blocks_session:N_blocks)*trials_block,'ytick',(0:ceil(M))) ;
set(findobj(gca,'Color','r'),'MarkerFaceColor','r');
%set(findobj(gca,'Color','g'),'MarkerFaceColor','g');
xlabel('Trial') ; ylabel('d''') ;

legend('Training order','Transfer order','Easy delta','Diff delta',...
      '1st order groups','2nd order groups','Location','NorthWest') ;

% %% Plot group d' at half temporal resolution
% idx2 = 2:2:24 ;
%
% g13easy2 = (g13easy(idx2-1) + g13easy(idx2))./2 ;
% g13hard2 = (g13hard(idx2-1) + g13hard(idx2))./2 ;
% g24easy2 = (g24easy(idx2-1) + g24easy(idx2))./2 ;
% g24hard2 = (g24hard(idx2-1) + g24hard(idx2))./2 ;
%
% globalT2 = (globalT(idx2-1) + globalT(idx2))./2 ;
%
% pre_test2  = (1:2)' ;   % pre-test blocks
% pre_tn2 = (2)' ;      % pre-test training order
% pre_tf2 = (1)' ;      % pre-test transfer order
% post_test2 = (9:10)' ; % post-test blocks
% post_tn2 = (9)' ;   % post-test training order
% post_tf2 = (10)' ;   % post-test transfer order
% tn2 = (2:9)' ;         % train in respective order
% tn_opp2 = (10:12)' ;    % train opposite order
%
% %%Plot the smoothed group-averaged d'
% clf
% M = 3.0 ;
% plot(-1,-1,'ks',-1,-1,'ko',-1,-1,'k-',-1,-1,'k--',...
%     -1,1,'rs',-1,1,'bo',...   % for the legend
%      globalT2(pre_tn2),g13easy2(pre_tn2),'rs',...
%      globalT2(pre_tn2),g13hard2(pre_tn2),'rs--',...
%      globalT2(pre_tf2),g13easy2(pre_tf2),'ro',...
%      globalT2(pre_tf2),g13hard2(pre_tf2),'ro--',...
%      globalT2(pre_tn2),g24easy2(pre_tn2),'bs',...
%      globalT2(pre_tn2),g24hard2(pre_tn2),'bs--',...
%      globalT2(pre_tf2),g24easy2(pre_tf2),'bo',...
%      globalT2(pre_tf2),g24hard2(pre_tf2),'bo--',...
%      globalT2(tn2),g13easy2(tn2),'rs-',...
%      globalT2(tn2),g13hard2(tn2),'rs--',...
%      globalT2(tn2),g24easy2(tn2),'bs-',...
%      globalT2(tn2),g24hard2(tn2),'bs--',...
%      globalT2(post_tn2),g13easy2(post_tn2),'rs',...
%      globalT2(post_tn2),g13hard2(post_tn2),'rs--',...
%      globalT2(post_tf2),g13easy2(post_tf2),'ro',...
%      globalT2(post_tf2),g13hard2(post_tf2),'ro--',...
%      globalT2(post_tn2),g24easy2(post_tn2),'bs',...
%      globalT2(post_tn2),g24hard2(post_tn2),'bs--',...
%      globalT2(post_tf2),g24easy2(post_tf2),'bo',...
%      globalT2(post_tf2),g24hard2(post_tf2),'bo--',...
%      globalT2(tn_opp2),g13easy2(tn_opp2),'ro-',...
%      globalT2(tn_opp2),g13hard2(tn_opp2),'ro--',...
%      globalT2(tn_opp2),g24easy2(tn_opp2),'bo-',...
%      globalT2(tn_opp2),g24hard2(tn_opp2),'bo--')
% hold on
% axis([0 N_trials 0 M]) ; grid on ;
% set(gca,'xtick',(0:blocks_session:N_blocks)*trials_block,'ytick',(0:ceil(M))) ;
% set(findobj(gca,'Color','r'),'MarkerFaceColor','r');
% %set(findobj(gca,'Color','g'),'MarkerFaceColor','g');
% xlabel('Trial') ; ylabel('d''') ;
% %title(sprintf('Group Average Pre/Post-Test, Train, and Train Opposite (10 vs 6 delta split)'));
% legend('Training order','Transfer order','Easy delta','Diff delta',...
%       '1st order groups','2nd order groups','Location','NorthWest') ;

Plot group d' for delta 11 (easy_dprime2)

gr = [H.group] ;
order1 = H(5).order2 ;
order2 = H(1).order2 ;


% Create a single averaged 28 x 1 matrix for gr 1+3 gr 2+4 and gr 1+2+3+4
% Group Averages all_dprime2
g13 = (mdpr2(:,1) + mdpr2(:,3)) / 2 ;
g24 = (mdpr2(:,2) + mdpr2(:,4)) / 2 ;
%g1234 = (mdpr2(:,1) + mdpr2(:,2) + mdpr2(:,3) + mdpr2(:,4)) / 4 ;

% Group Averages easy_dprime2 and hard_dprime2
g13easy = (measy2(:,1) + measy2(:,3)) / 2 ;
g13hard = (mhard2(:,1) + mhard2(:,3)) / 2 ;
g24easy = (measy2(:,2) + measy2(:,4)) / 2 ;
g24hard = (mhard2(:,2) + mhard2(:,4)) / 2 ;

pre_test  = (1:4)' ;   % pre-test blocks
pre_tn = (3:4)' ;      % pre-test training order
pre_tf = (1:2)' ;      % pre-test transfer order
post_test = (17:20)' ; % post-test blocks
post_tn = (17:18)' ;   % post-test training order
post_tf = (19:20)' ;   % post-test transfer order
tn = (3:18)' ;         % train in respective order
tn_opp = (19:24)' ;    % train opposite order

globalT = (bl-.5)*trials_block ;

% Loop to find the mean d_prime for all subjects in block2
for k=1:N_sbj
x(k) = mean(H(k).easy_dprime2);
end;

% Loop that subtracts the mean from each point in all_dprime2
% Sets all sbj mean dprime to zero
% 20 subjects x 24 adjusted dprime values
for k=1:20
U(k,:) = H(k).easy_dprime2(:) - x(k);
end

% Separate into train first and train second groups / create index
gr = [H.group];
gr = gr';
idx1 = find(gr==1 | gr==3);
idx2 = find(gr==2 | gr==4);
U1 = U(idx1,:);
U2 = U(idx2,:);


% Loop to compute standard error for each block for group1
for k=1:24
E(k) = std(U1(:,k));
ER1(k) = E(k)/sqrt(10);
end

% Loop to compute standard error for each block for group2
for k=1:24
E(k) = std(U2(:,k));
ER2(k) = E(k)/sqrt(10);
end

JTR = 10 ;   % horizontal jitter [in trials] for visibility
plot(-1,-1,'ks',-1,-1,'ko',-1,1,'rs',-1,1,'bo',...   % for the legend
     globalT(pre_tn)-JTR,g13easy(pre_tn),'rs',...
     globalT(pre_tf)-JTR,g13easy(pre_tf),'ro',...
     globalT(pre_tn)+JTR,g24easy(pre_tn),'bs',...
     globalT(pre_tf)+JTR,g24easy(pre_tf),'bo',...
     globalT(tn)-JTR,g13easy(tn),'rs-',...
     globalT(tn)+JTR,g24easy(tn),'bs-',...
     globalT(post_tn)-JTR,g13easy(post_tn),'rs',...
     globalT(post_tf)-JTR,g13easy(post_tf),'ro',...
     globalT(post_tn)+JTR,g24easy(post_tn),'bs',...
     globalT(post_tf)+JTR,g24easy(post_tf),'bo',...
     globalT(tn_opp)-JTR,g13easy(tn_opp),'ro-',...
     globalT(tn_opp)+JTR,g24easy(tn_opp),'bo-') ;
hold on ; errorbar1(globalT-JTR,g13easy,ER1,'rn'); hold off
hold on ; errorbar1(globalT+JTR,g24easy,ER2,'bn'); hold off

M = 2.5 ;
axis([0 N_trials 0 M]) ; grid on ;
set(gca,'xtick',(0:blocks_session:N_blocks)*trials_block,'ytick',(0:ceil(M))) ;
set(findobj(gca,'Color','r'),'MarkerFaceColor','r');
%set(findobj(gca,'Color','g'),'MarkerFaceColor','g');
xlabel('Trial') ; ylabel('d''') ;
%title(sprintf('Group Average Pre/Post-Test, Train, and Train Opposite'));
legend('Training order','Transfer order','1st order groups','2nd order groups','Location','NorthWest') ;


% %% Plot group d' at half temporal resolution
% idx2 = 2:2:24 ;
%
% g13easy2 = (g13easy(idx2-1) + g13easy(idx2))./2 ;
% g13hard2 = (g13hard(idx2-1) + g13hard(idx2))./2 ;
% g24easy2 = (g24easy(idx2-1) + g24easy(idx2))./2 ;
% g24hard2 = (g24hard(idx2-1) + g24hard(idx2))./2 ;
%
% globalT2 = (globalT(idx2-1) + globalT(idx2))./2 ;
%
% pre_test2  = (1:2)' ;   % pre-test blocks
% pre_tn2 = (2)' ;      % pre-test training order
% pre_tf2 = (1)' ;      % pre-test transfer order
% post_test2 = (9:10)' ; % post-test blocks
% post_tn2 = (9)' ;   % post-test training order
% post_tf2 = (10)' ;   % post-test transfer order
% tn2 = (2:9)' ;         % train in respective order
% tn_opp2 = (10:12)' ;    % train opposite order
%
% %%Plot the smoothed group-averaged d'
% clf
% M = 3.0 ;
% plot(-1,-1,'ks',-1,-1,'ko',-1,-1,'k-',-1,-1,'k--',...
%     -1,1,'rs',-1,1,'bo',...   % for the legend
%      globalT2(pre_tn2),g13easy2(pre_tn2),'rs',...
%      globalT2(pre_tn2),g13hard2(pre_tn2),'rs--',...
%      globalT2(pre_tf2),g13easy2(pre_tf2),'ro',...
%      globalT2(pre_tf2),g13hard2(pre_tf2),'ro--',...
%      globalT2(pre_tn2),g24easy2(pre_tn2),'bs',...
%      globalT2(pre_tn2),g24hard2(pre_tn2),'bs--',...
%      globalT2(pre_tf2),g24easy2(pre_tf2),'bo',...
%      globalT2(pre_tf2),g24hard2(pre_tf2),'bo--',...
%      globalT2(tn2),g13easy2(tn2),'rs-',...
%      globalT2(tn2),g13hard2(tn2),'rs--',...
%      globalT2(tn2),g24easy2(tn2),'bs-',...
%      globalT2(tn2),g24hard2(tn2),'bs--',...
%      globalT2(post_tn2),g13easy2(post_tn2),'rs',...
%      globalT2(post_tn2),g13hard2(post_tn2),'rs--',...
%      globalT2(post_tf2),g13easy2(post_tf2),'ro',...
%      globalT2(post_tf2),g13hard2(post_tf2),'ro--',...
%      globalT2(post_tn2),g24easy2(post_tn2),'bs',...
%      globalT2(post_tn2),g24hard2(post_tn2),'bs--',...
%      globalT2(post_tf2),g24easy2(post_tf2),'bo',...
%      globalT2(post_tf2),g24hard2(post_tf2),'bo--',...
%      globalT2(tn_opp2),g13easy2(tn_opp2),'ro-',...
%      globalT2(tn_opp2),g13hard2(tn_opp2),'ro--',...
%      globalT2(tn_opp2),g24easy2(tn_opp2),'bo-',...
%      globalT2(tn_opp2),g24hard2(tn_opp2),'bo--')
% hold on
%
% axis([0 N_trials 0 M]) ; grid on ;
% set(gca,'xtick',(0:blocks_session:N_blocks)*trials_block,'ytick',(0:ceil(M))) ;
% set(findobj(gca,'Color','r'),'MarkerFaceColor','r');
% %set(findobj(gca,'Color','g'),'MarkerFaceColor','g');
% xlabel('Trial') ; ylabel('d''') ;
% %title(sprintf('Group Average Pre/Post-Test, Train, and Train Opposite (10 vs 6 delta split)'));
% legend('Training order','Transfer order','Easy delta','Diff delta',...
%       '1st order groups','2nd order groups','Location','NorthWest') ;

Plot group d' for delta 8 (hard_dprime2)

gr = [H.group] ;
order1 = H(5).order2 ;
order2 = H(1).order2 ;


% Create a single averaged 28 x 1 matrix for gr 1+3 gr 2+4 and gr 1+2+3+4
% Group Averages all_dprime2
g13 = (mdpr2(:,1) + mdpr2(:,3)) / 2 ;
g24 = (mdpr2(:,2) + mdpr2(:,4)) / 2 ;
%g1234 = (mdpr2(:,1) + mdpr2(:,2) + mdpr2(:,3) + mdpr2(:,4)) / 4 ;

% Group Averages easy_dprime2 and hard_dprime2
g13easy = (measy2(:,1) + measy2(:,3)) / 2 ;
g13hard = (mhard2(:,1) + mhard2(:,3)) / 2 ;
g24easy = (measy2(:,2) + measy2(:,4)) / 2 ;
g24hard = (mhard2(:,2) + mhard2(:,4)) / 2 ;

pre_test  = (1:4)' ;   % pre-test blocks
pre_tn = (3:4)' ;      % pre-test training order
pre_tf = (1:2)' ;      % pre-test transfer order
post_test = (17:20)' ; % post-test blocks
post_tn = (17:18)' ;   % post-test training order
post_tf = (19:20)' ;   % post-test transfer order
tn = (3:18)' ;         % train in respective order
tn_opp = (19:24)' ;    % train opposite order

globalT = (bl-.5)*trials_block ;

% Loop to find the mean d_prime for all subjects in block2
for k=1:N_sbj
x(k) = mean(H(k).hard_dprime2);
end;

% Loop that subtracts the mean from each point in all_dprime2
% Sets all sbj mean dprime to zero
% 20 subjects x 24 adjusted dprime values
for k=1:20
U(k,:) = H(k).hard_dprime2(:) - x(k);
end

% Separate into train first and train second groups / create index
gr = [H.group];
gr = gr';
idx1 = find(gr==1 | gr==3);
idx2 = find(gr==2 | gr==4);
U1 = U(idx1,:);
U2 = U(idx2,:);


% Loop to compute standard error for each block for group1
for k=1:24
E(k) = std(U1(:,k));
ER1(k) = E(k)/sqrt(10);
end

% Loop to compute standard error for each block for group2
for k=1:24
E(k) = std(U2(:,k));
ER2(k) = E(k)/sqrt(10);
end

JTR = 10 ;   % horizontal jitter [in trials] for visibility
plot(-1,-1,'ks',-1,-1,'ko',-1,1,'rs',-1,1,'bo',...   % for the legend
     globalT(pre_tn)-JTR,g13hard(pre_tn),'rs',...
     globalT(pre_tf)-JTR,g13hard(pre_tf),'ro',...
     globalT(pre_tn)+JTR,g24hard(pre_tn),'bs',...
     globalT(pre_tf)+JTR,g24hard(pre_tf),'bo',...
     globalT(tn)-JTR,g13hard(tn),'rs-',...
     globalT(tn)+JTR,g24hard(tn),'bs-',...
     globalT(post_tn)-JTR,g13hard(post_tn),'rs',...
     globalT(post_tf)-JTR,g13hard(post_tf),'ro',...
     globalT(post_tn)+JTR,g24hard(post_tn),'bs',...
     globalT(post_tf)+JTR,g24hard(post_tf),'bo',...
     globalT(tn_opp)-JTR,g13hard(tn_opp),'ro-',...
     globalT(tn_opp)+JTR,g24hard(tn_opp),'bo-') ;
hold on ; errorbar1(globalT-JTR,g13hard,ER1,'rn'); hold off
hold on ; errorbar1(globalT+JTR,g24hard,ER2,'bn'); hold off

M = 2.5 ;
axis([0 N_trials 0 M]) ; grid on ;
set(gca,'xtick',(0:blocks_session:N_blocks)*trials_block,'ytick',(0:ceil(M))) ;
set(findobj(gca,'Color','r'),'MarkerFaceColor','r');
%set(findobj(gca,'Color','g'),'MarkerFaceColor','g');
xlabel('Trial') ; ylabel('d''') ;
%title(sprintf('Group Average Pre/Post-Test, Train, and Train Opposite'));
legend('Training order','Transfer order','1st order groups','2nd order groups','Location','NorthWest') ;


% %% Plot group d' at half temporal resolution
% idx2 = 2:2:24 ;
%
% g13easy2 = (g13easy(idx2-1) + g13easy(idx2))./2 ;
% g13hard2 = (g13hard(idx2-1) + g13hard(idx2))./2 ;
% g24easy2 = (g24easy(idx2-1) + g24easy(idx2))./2 ;
% g24hard2 = (g24hard(idx2-1) + g24hard(idx2))./2 ;
%
% globalT2 = (globalT(idx2-1) + globalT(idx2))./2 ;
%
% pre_test2  = (1:2)' ;   % pre-test blocks
% pre_tn2 = (2)' ;      % pre-test training order
% pre_tf2 = (1)' ;      % pre-test transfer order
% post_test2 = (9:10)' ; % post-test blocks
% post_tn2 = (9)' ;   % post-test training order
% post_tf2 = (10)' ;   % post-test transfer order
% tn2 = (2:9)' ;         % train in respective order
% tn_opp2 = (10:12)' ;    % train opposite order
%
% %%Plot the smoothed group-averaged d'
% clf
% M = 3.0 ;
% plot(-1,-1,'ks',-1,-1,'ko',-1,-1,'k-',-1,-1,'k--',...
%     -1,1,'rs',-1,1,'bo',...   % for the legend
%      globalT2(pre_tn2),g13easy2(pre_tn2),'rs',...
%      globalT2(pre_tn2),g13hard2(pre_tn2),'rs--',...
%      globalT2(pre_tf2),g13easy2(pre_tf2),'ro',...
%      globalT2(pre_tf2),g13hard2(pre_tf2),'ro--',...
%      globalT2(pre_tn2),g24easy2(pre_tn2),'bs',...
%      globalT2(pre_tn2),g24hard2(pre_tn2),'bs--',...
%      globalT2(pre_tf2),g24easy2(pre_tf2),'bo',...
%      globalT2(pre_tf2),g24hard2(pre_tf2),'bo--',...
%      globalT2(tn2),g13easy2(tn2),'rs-',...
%      globalT2(tn2),g13hard2(tn2),'rs--',...
%      globalT2(tn2),g24easy2(tn2),'bs-',...
%      globalT2(tn2),g24hard2(tn2),'bs--',...
%      globalT2(post_tn2),g13easy2(post_tn2),'rs',...
%      globalT2(post_tn2),g13hard2(post_tn2),'rs--',...
%      globalT2(post_tf2),g13easy2(post_tf2),'ro',...
%      globalT2(post_tf2),g13hard2(post_tf2),'ro--',...
%      globalT2(post_tn2),g24easy2(post_tn2),'bs',...
%      globalT2(post_tn2),g24hard2(post_tn2),'bs--',...
%      globalT2(post_tf2),g24easy2(post_tf2),'bo',...
%      globalT2(post_tf2),g24hard2(post_tf2),'bo--',...
%      globalT2(tn_opp2),g13easy2(tn_opp2),'ro-',...
%      globalT2(tn_opp2),g13hard2(tn_opp2),'ro--',...
%      globalT2(tn_opp2),g24easy2(tn_opp2),'bo-',...
%      globalT2(tn_opp2),g24hard2(tn_opp2),'bo--')
% hold on
% axis([0 N_trials 0 M]) ; grid on ;
% set(gca,'xtick',(0:blocks_session:N_blocks)*trials_block,'ytick',(0:ceil(M))) ;
% set(findobj(gca,'Color','r'),'MarkerFaceColor','r');
% %set(findobj(gca,'Color','g'),'MarkerFaceColor','g');
% xlabel('Trial') ; ylabel('d''') ;
% %title(sprintf('Group Average Pre/Post-Test, Train, and Train Opposite (10 vs 6 delta split)'));
% legend('Training order','Transfer order','Easy delta','Diff delta',...
%       '1st order groups','2nd order groups','Location','NorthWest') ;

Plot group d' for delta 11 (easy_dprime2)

gr = [H.group] ;
order1 = H(5).order2 ;
order2 = H(1).order2 ;


% Create a single averaged 28 x 1 matrix for gr 1+3 gr 2+4 and gr 1+2+3+4
% Group Averages all_dprime2
g13 = (mdpr2(:,1) + mdpr2(:,3)) / 2 ;
g24 = (mdpr2(:,2) + mdpr2(:,4)) / 2 ;
%g1234 = (mdpr2(:,1) + mdpr2(:,2) + mdpr2(:,3) + mdpr2(:,4)) / 4 ;

% Group Averages easy_dprime2 and hard_dprime2
g13easy = (measy2(:,1) + measy2(:,3)) / 2 ;
g13hard = (mhard2(:,1) + mhard2(:,3)) / 2 ;
g24easy = (measy2(:,2) + measy2(:,4)) / 2 ;
g24hard = (mhard2(:,2) + mhard2(:,4)) / 2 ;

pre_test  = (1:4)' ;   % pre-test blocks
pre_tn = (3:4)' ;      % pre-test training order
pre_tf = (1:2)' ;      % pre-test transfer order
post_test = (17:20)' ; % post-test blocks
post_tn = (17:18)' ;   % post-test training order
post_tf = (19:20)' ;   % post-test transfer order
tn = (3:18)' ;         % train in respective order
tn_opp = (19:24)' ;    % train opposite order

globalT = (bl-.5)*trials_block ;

% Loop to find the mean d_prime for all subjects in block2
for k=1:N_sbj
x(k) = mean(H(k).easy_dprime2);
end;

% Loop that subtracts the mean from each point in all_dprime2
% Sets all sbj mean dprime to zero
% 20 subjects x 24 adjusted dprime values
for k=1:20
U(k,:) = H(k).easy_dprime2(:) - x(k);
end

% Separate into train first and train second groups / create index
gr = [H.group];
gr = gr';
idx1 = find(gr==1 | gr==3);
idx2 = find(gr==2 | gr==4);
U1 = U(idx1,:);
U2 = U(idx2,:);


% Loop to compute standard error for each block for group1
for k=1:24
E(k) = std(U1(:,k));
ER1(k) = E(k)/sqrt(10);
end

% Loop to compute standard error for each block for group2
for k=1:24
E(k) = std(U2(:,k));
ER2(k) = E(k)/sqrt(10);
end

JTR = 10 ;   % horizontal jitter [in trials] for visibility
plot(-1,-1,'ks',-1,-1,'ko',-1,1,'rs',-1,1,'bo',...   % for the legend
     globalT(pre_tn)-JTR,g13easy(pre_tn),'rs',...
     globalT(pre_tf)-JTR,g13easy(pre_tf),'ro',...
     globalT(pre_tn)+JTR,g24easy(pre_tn),'bs',...
     globalT(pre_tf)+JTR,g24easy(pre_tf),'bo',...
     globalT(tn)-JTR,g13easy(tn),'rs-',...
     globalT(tn)+JTR,g24easy(tn),'bs-',...
     globalT(post_tn)-JTR,g13easy(post_tn),'rs',...
     globalT(post_tf)-JTR,g13easy(post_tf),'ro',...
     globalT(post_tn)+JTR,g24easy(post_tn),'bs',...
     globalT(post_tf)+JTR,g24easy(post_tf),'bo',...
     globalT(tn_opp)-JTR,g13easy(tn_opp),'ro-',...
     globalT(tn_opp)+JTR,g24easy(tn_opp),'bo-') ;
hold on ; errorbar1(globalT-JTR,g13easy,ER1,'rn'); hold off
hold on ; errorbar1(globalT+JTR,g24easy,ER2,'bn'); hold off

hold on ;

% Plot group d' for delta 8 (hard_dprime2)
gr = [H.group] ;
order1 = H(5).order2 ;
order2 = H(1).order2 ;


% Create a single averaged 28 x 1 matrix for gr 1+3 gr 2+4 and gr 1+2+3+4
% Group Averages all_dprime2
g13 = (mdpr2(:,1) + mdpr2(:,3)) / 2 ;
g24 = (mdpr2(:,2) + mdpr2(:,4)) / 2 ;
%g1234 = (mdpr2(:,1) + mdpr2(:,2) + mdpr2(:,3) + mdpr2(:,4)) / 4 ;

% Group Averages easy_dprime2 and hard_dprime2
g13easy = (measy2(:,1) + measy2(:,3)) / 2 ;
g13hard = (mhard2(:,1) + mhard2(:,3)) / 2 ;
g24easy = (measy2(:,2) + measy2(:,4)) / 2 ;
g24hard = (mhard2(:,2) + mhard2(:,4)) / 2 ;

pre_test  = (1:4)' ;   % pre-test blocks
pre_tn = (3:4)' ;      % pre-test training order
pre_tf = (1:2)' ;      % pre-test transfer order
post_test = (17:20)' ; % post-test blocks
post_tn = (17:18)' ;   % post-test training order
post_tf = (19:20)' ;   % post-test transfer order
tn = (3:18)' ;         % train in respective order
tn_opp = (19:24)' ;    % train opposite order

globalT = (bl-.5)*trials_block ;

% Loop to find the mean d_prime for all subjects in block2
for k=1:N_sbj
x(k) = mean(H(k).hard_dprime2);
end;

% Loop that subtracts the mean from each point in all_dprime2
% Sets all sbj mean dprime to zero
% 20 subjects x 24 adjusted dprime values
for k=1:20
U(k,:) = H(k).hard_dprime2(:) - x(k);
end

% Separate into train first and train second groups / create index
gr = [H.group];
gr = gr';
idx1 = find(gr==1 | gr==3);
idx2 = find(gr==2 | gr==4);
U1 = U(idx1,:);
U2 = U(idx2,:);


% Loop to compute standard error for each block for group1
for k=1:24
E(k) = std(U1(:,k));
ER1(k) = E(k)/sqrt(10);
end

% Loop to compute standard error for each block for group2
for k=1:24
E(k) = std(U2(:,k));
ER2(k) = E(k)/sqrt(10);
end

JTR = 20 ;   % horizontal jitter [in trials] for visibility
plot(-1,-1,'ks',-1,-1,'ko',-1,1,'rs',-1,1,'bo',...   % for the legend
     globalT(pre_tn)-JTR,g13hard(pre_tn),'rs',...
     globalT(pre_tf)-JTR,g13hard(pre_tf),'ro',...
     globalT(pre_tn)+JTR,g24hard(pre_tn),'bs',...
     globalT(pre_tf)+JTR,g24hard(pre_tf),'bo',...
     globalT(tn)-JTR,g13hard(tn),'rs-',...
     globalT(tn)+JTR,g24hard(tn),'bs-',...
     globalT(post_tn)-JTR,g13hard(post_tn),'rs',...
     globalT(post_tf)-JTR,g13hard(post_tf),'ro',...
     globalT(post_tn)+JTR,g24hard(post_tn),'bs',...
     globalT(post_tf)+JTR,g24hard(post_tf),'bo',...
     globalT(tn_opp)-JTR,g13hard(tn_opp),'ro-',...
     globalT(tn_opp)+JTR,g24hard(tn_opp),'bo-') ;
hold on ; errorbar1(globalT-JTR,g13hard,ER1,'rn'); hold off
hold on ; errorbar1(globalT+JTR,g24hard,ER2,'bn'); hold off

M = 2.5 ;
axis([0 N_trials 0 M]) ; grid on ;
set(gca,'xtick',(0:blocks_session:N_blocks)*trials_block,'ytick',(0:ceil(M))) ;
set(findobj(gca,'Color','r'),'MarkerFaceColor','r');
%set(findobj(gca,'Color','g'),'MarkerFaceColor','g');
xlabel('Trial') ; ylabel('d''') ;
%title(sprintf('Group Average Pre/Post-Test, Train, and Train Opposite'));
legend('Training order','Transfer order','1st order groups','2nd order groups','Location','NorthWest') ;

Sbj Order Learning Effect Scatter plot

pre_test  = (1:4)' ;   % pre-test blocks
pre_tn = (3:4)' ;      % pre-test training order
pre_tf = (1:2)' ;      % pre-test transfer order
post_test = (17:20)' ; % post-test blocks
post_tn = (17:18)' ;   % post-test training order
post_tf = (19:20)' ;   % post-test transfer order
tn = (3:18)' ;         % train in respective order
tn_opp = (19:24)' ;    % train opposite order

clf ; hold on ;
for j=1:N_sbj
    LE1 = H(j).all_dprime2;
    base_tf = sum(LE1(pre_tf))/2;
    base_tn = sum(LE1(pre_tn))/2;
    fin_tf = sum(LE1(post_tf))/2;
    fin_tn = sum(LE1(post_tn))/2;
    % Calculate Learning effects
    tf_LE = (fin_tf-base_tf)/base_tf;
    tn_LE = (fin_tn-base_tn)/base_tn;
    if (gr(j)==1 || gr(j)==3) ; symb = 'rs' ; else symb = 'bo' ; end
    plot(tn_LE,tf_LE,symb);
end
hold off ;
m=-.8 ; M = 3.5 ;
axis([m M m M]) ; grid on ; axis square ;
xlabel('Training learning effect'); ylabel('Transfer Learning Effect');
title(sprintf('Group Learning Effect Scatterplot'));
refline(1,0) ;

Group Order Learning Effect bar plot

%pre_tn = (3:4)' ;      % pre-test training order
pre_tn = (3)';
%pre_tf = (1:2)' ;      % pre-test transfer order
pre_tf = (1)';
post_test = (17:20)' ; % post-test blocks
%post_tn = (17:18)' ;   % post-test training order
post_tn = (17)';
%post_tf = (19:20)' ;   % post-test transfer order
post_tf = (19)';
tn = (3:18)' ;         % train in respective order
tn_opp = (20:24)' ;    % train opposite order
g13 = (mdpr2(:,1)+ mdpr2(:,3))/2 ;
g24 = (mdpr2(:,2) + mdpr2(:,4)) / 2 ;
clf ; hold on ;

base_tf1 = mean(g13(pre_tf));
base_tn1 = mean(g13(pre_tn));
fin_tf1 = mean(g13(post_tf));
fin_tn1 = mean(g13(post_tn));
% Calculate Learning effects
tf_LE1 = (fin_tf1-base_tf1)/base_tf1;
tn_LE1 = (fin_tn1-base_tn1)/base_tn1;

base_tf2 = mean(g24(pre_tf));
base_tn2 = mean(g24(pre_tn));
fin_tf2 = mean(g24(post_tf));
fin_tn2 = mean(g24(post_tn));
% Calculate Learning effects
tf_LE2 = (fin_tf2-base_tf2)/base_tf2;
tn_LE2 = (fin_tn2-base_tn2)/base_tn2;

hold on
y = [tn_LE1 tf_LE1
     tn_LE2 tf_LE2];
bar(y,1);
ylabel('Learning Effect');
hold off
YLIM([0 .8]);
set(gca,'xtick',(0:1:4),'ytick',(0:.2:2));%'XTickLabel',{'second tn second tf','first tn   first tf'});


% hold off ;
% m=0 ; M = 1 ;
% axis([m M m M]) ; grid on ; axis square ;
% xlabel('Training learning effect'); ylabel('Transfer Learning Effect');
% title(sprintf('Group Learning Effect Scatterplot'));
% refline(1,0) ;

Warning: Could not find an exact (case-sensitive) match for 'YLIM'.
/Applications/MATLAB_R2008a/toolbox/matlab/graph3d/ylim.m is a case-insensitive match and
will be used instead.
You can improve the performance of your code by using exact
name matches and we therefore recommend that you update your
usage accordingly. Alternatively, you can disable this warning using
warning('off','MATLAB:dispatcher:InexactCaseMatch').
This warning will become an error in future releases.

Group Order Learning Effect bar plot

pre_tn = (3:4)' ;      % pre-test training order
pre_tf = (1:2)' ;      % pre-test transfer order
post_test = (17:20)' ; % post-test blocks
post_tn = (17:18)' ;   % post-test training order
post_tf = (19:20)' ;   % post-test transfer order
tn = (3:18)' ;         % train in respective order
tn_opp = (19:24)' ;    % train opposite order
g13 = (mdpr2(:,1)+ mdpr2(:,3))/2 ;
g24 = (mdpr2(:,2) + mdpr2(:,4)) / 2 ;
clf ; hold on ;

base_tf1 = mean(g13(pre_tf));
base_tn1 = mean(g13(pre_tn));
fin_tf1 = mean(g13(post_tf));
fin_tn1 = mean(g13(post_tn));
% Calculate Learning effects
tf_LE1 = (fin_tf1-base_tf1)/base_tf1;
tn_LE1 = (fin_tn1-base_tn1)/base_tn1;

base_tf2 = mean(g24(pre_tf));
base_tn2 = mean(g24(pre_tn));
fin_tf2 = mean(g24(post_tf));
fin_tn2 = mean(g24(post_tn));
% Calculate Learning effects
tf_LE2 = (fin_tf2-base_tf2)/base_tf2;
tn_LE2 = (fin_tn2-base_tn2)/base_tn2;

hold on
y = [tn_LE1 tf_LE1
     tn_LE2 tf_LE2];
bar(y,1);
ylabel('Learning Effect');
hold off
YLIM([0 .8]);
set(gca,'xtick',(0:1:4),'ytick',(0:.2:2));%'XTickLabel',{'second tn second tf','first tn   first tf'});


% hold off ;
% m=0 ; M = 1 ;
% axis([m M m M]) ; grid on ; axis square ;
% xlabel('Training learning effect'); ylabel('Transfer Learning Effect');
% title(sprintf('Group Learning Effect Scatterplot'));
% refline(1,0) ;

Main effect of Order on dprime:

l_1st2nd = {'Gr 1+3: 1st','Gr 2+4: 2nd'} ; % legend
Ss_1st2nd = cell(2,1) ;
for k = 1:2
    if (k==1)   % Train order
        Ss_1st2nd{k} = find(gr==1 | gr==3) ;
    else %(k==2)  Train 2nd order
        Ss_1st2nd{k} = find(gr==2 | gr==4) ;
    end
    mdpr2_1st2nd(:,k) = mean(dpr2(:,Ss_1st2nd{k}),2) ;    % average over subjects
    mcrit2_1st2nd(:,k) = mean(crit2(:,Ss_1st2nd{k}),2) ;
    measy2_1st2nd(:,k) = mean(easy2(:,Ss_1st2nd{k}),2) ;
    mhard2_1st2nd(:,k) = mean(hard2(:,Ss_1st2nd{k}),2) ;
    mRT2_1st2nd(:,k) = mean(medianRT2(:,Ss_1st2nd{k}),2) ;
end
assert(isempty(intersect(Ss_1st2nd{1},Ss_1st2nd{2}))) ;   % Sanity check

clf
M = 3.5 ;
plot(globalT,mdpr2_1st2nd(:,1),'rs-',...
     globalT,mdpr2_1st2nd(:,2),'bo-') ;
axis([0 N_trials 0 M]) ; grid on ;
set(findobj(gca,'Color','r'),'MarkerFaceColor','r') ;
set(gca,'xtick',(0:blocks_session:N_blocks)*trials_block,'ytick',(0:ceil(M))) ;
xlabel('Trial') ; ylabel('d''') ;
title(sprintf('%s, %d Ss; %s, %d Ss',...
      l_1st2nd{1},length(Ss_1st2nd{1}),l_1st2nd{2},length(Ss_1st2nd{2}))) ;
legend(l_1st2nd,'location','NorthEast') ;

Group d'prime error plot

%{
% Loop to find the mean d_prime for all subjects in block2
for k=1:N_sbj
x(k) = mean(H(k).all_dprime2);
end;

% Loop that subtracts the mean from each point in all_dprime2
% Sets all sbj mean dprime to zero
% 20 subjects x 24 adjusted dprime values
for k=1:20
U(k,:) = H(k).all_dprime2(:) - x(k);
end

% Separate into train first and train second groups / create index
gr = [H.group];
gr = gr';
idx1 = find(gr==1 | gr==3);
idx2 = find(gr==2 | gr==4);
U1 = U(idx1,:);
U2 = U(idx2,:);


% Loop to compute standard error for each block for group1
for k=1:24
E(k) = std(U1(:,k));
ER1(k) = E(k)/sqrt(10);
end

% Loop to compute standard error for each block for group2
for k=1:24
E(k) = std(U2(:,k));
ER2(k) = E(k)/sqrt(10);
end


clf
M = 2.5 ;
fE = errorbar(globalT-JTR,mdpr2_1st2nd(:,1),ER1.*norminv(.95),'rs-');
hold on
%fE = errorbar([globalT-10;globalT+10],mdpr2_1st2nd(:),[ER1,ER2].*norminv(.95),'rs-');
sE = errorbar(globalT+JTR,mdpr2_1st2nd(:,2),ER2.*norminv(.95),'bo-');
hold off
oldAx = axis;
%removeErrorbarends(fE);
%removeErrorbarends(sE);
axis(oldAx);
axis([0 N_trials 0 M]) ; grid on ;
set(findobj(gca,'Color','r'),'MarkerFaceColor','r') ;
set(gca,'xtick',(0:blocks_session:N_blocks)*trials_block,'ytick',(0:ceil(M))) ;
%}

Group criterion plot:

clf
m = .5 ;
plot(globalT,mcrit2_1st2nd(:,1),'rs-',...
     globalT,mcrit2_1st2nd(:,2),'bo-') ;
axis([0 N_trials -m m]) ; grid on ;
set(findobj(gca,'Color','r'),'MarkerFaceColor','r') ;
set(gca,'xtick',(0:blocks_session:N_blocks)*trials_block) ;
xlabel('Trial') ; ylabel('criterion') ;
title(sprintf('%s, %d Ss; %s, %d Ss',...
      l_1st2nd{1},length(Ss_1st2nd{1}),l_1st2nd{2},length(Ss_1st2nd{2}))) ;
legend(l_1st2nd,'location','NorthEast') ;

% sprintf('%s, %d Ss; %s, %d Ss; %s, %d Ss; %s, %d Ss',...
%       l{1},length(Ss{1}),l{2},length(Ss{2}),...
%       l{3},length(Ss{3}),l{4},length(Ss{4}))) ;
% legend(l,'location','NorthWest') ;


% %% Easy Delta group plot
%
% clf
% M = 3.5 ;
% plot(globalT(train1steasy),g13easy(train1steasy),'rs-',...
%      globalT(train2ndeasy),g24easy(train2ndeasy),'bo-') ;
% axis([0 N_trials 0 M]) ; grid on ;
% set(findobj(gca,'Color','r'),'MarkerFaceColor','r') ;
% set(gca,'xtick',(0:blocks_session:N_blocks)*trials_block,'ytick',(0:ceil(M))) ;
% xlabel('Trial') ; ylabel('d''') ;
% title(sprintf('%s, %d Ss; %s, %d Ss',...
%        l_1st2nd{1},length(Ss_1st2nd{1}),l_1st2nd{2},length(Ss_1st2nd{2})));
% legend(l_1st2nd,'location','SouthEast') ;
%
%
% %% Hard Delta group plot
% clf
% M = 3.5 ;
% plot(globalT(trainhard),g13hard(trainhard),'rs--',...
%      globalT(trainhard),g24hard(trainhard),'bo--')
% axis([0 N_trials 0 M]) ; grid on ;
% set(findobj(gca,'Color','r'),'MarkerFaceColor','r') ;
% set(gca,'xtick',(0:blocks_session:N_blocks)*trials_block,'ytick',(0:ceil(M))) ;
% xlabel('Trial') ; ylabel('d''') ;
% title(sprintf('%s, %d Ss; %s, %d Ss',...
%        l_1st2nd{1},length(Ss_1st2nd{1}),l_1st2nd{2},length(Ss_1st2nd{2})));
% legend(l_1st2nd,'location','SouthEast') ;

Group RT plot

clf
M = 1000 ;  % msec
m = 300 ;

plot(globalT,mRT2(:,1),'rs-',...
     globalT,mRT2(:,2),'bs-',...
     globalT,mRT2(:,3),'r^-',...
     globalT,mRT2(:,4),'b^-') ;
axis([0 N_trials m M]) ; grid on ;
set(findobj(gca,'Color','r'),'MarkerFaceColor','r') ;
set(gca,'xtick',(0:blocks_session:N_blocks)*trials_block) ;
xlabel('Trial') ; ylabel('mean(median(RT))') ;
title(sprintf('%s, %d Ss; %s, %d Ss; %s, %d Ss; %s, %d Ss',...
      l{1},length(Ss{1}),l{2},length(Ss{2}),...
      l{3},length(Ss{3}),l{4},length(Ss{4}))) ;
legend(l,'location','SouthWest') ;


% %% Plot all RT Distributions
% sbj = 25;
% clf; hold on;
% for k =1:sbj
% subplot(5,5,k);
% x = D(1,k).data(:,15);
% idx = find(x<4000);
% RT = x(idx,:);
% hist(RT,50);
% axis([0 3500 0 2000]);
% xlabel('time ms', 'fontsize', 8);
% %set(gca,'fontsize',14);
% end

% %% Individual RT Plot
% % for individual plots replace k with value 1-25
% x = D(1,5).data(:,15);
% idx = find(x<4000);
% RT = x(idx,:);
% hist(RT,90);
% axis([0 3000 0 1000]);
% h = findobj(gca,'Type','patch');
% set(h,'FaceColor','b','EdgeColor','w');
% xlabel('time ms', 'fontsize', 16); ylabel('frequency','fontsize',16);
% title('Sbj 132','fontsize',18);
% set(gca,'fontsize',14);

Main effect of Order on RT

clf
M = 1100 ;  % msec
m = 500 ;
plot(globalT,mRT2_1st2nd(:,1),'rs-',...
     globalT,mRT2_1st2nd(:,2),'bo-') ;
axis([0 N_trials m M]) ; grid on ;
set(findobj(gca,'Color','r'),'MarkerFaceColor','r') ;
set(gca,'xtick',(0:blocks_session:N_blocks)*trials_block) ;
xlabel('Trial') ; ylabel('mean(median(RT))') ;
title(sprintf('%s, %d Ss; %s, %d Ss',...
      l_1st2nd{1},length(Ss_1st2nd{1}),l_1st2nd{2},length(Ss_1st2nd{2}))) ;
legend(l_1st2nd,'location','NorthEast') ;

Adjusted group RT plot

Adjust all RTs so that the two curves have the same mean on
the pretest, which is identical for both groups.

clf
baseRT1 = mean(mRT2_1st2nd(pre_test,1)) ;
baseRT2 = mean(mRT2_1st2nd(pre_test,2)) ;
baseRT_adjustment = [baseRT2-baseRT1 , baseRT1-baseRT2] ./ 2 ;
fprintf('\nMean RT(Gr1+3,day1)=%.0f, Mean RT(Gr2+4,day1)=%.0f, adjustment=2*%.0f msec.\n',...
        baseRT1,baseRT2,baseRT_adjustment(1)) ;

mRT2_1st2nd_adj = mRT2_1st2nd + repmat(baseRT_adjustment,size(mRT2_1st2nd,1),1) ;

plot(globalT,mRT2_1st2nd_adj(:,1),'rs-',...
     globalT,mRT2_1st2nd_adj(:,2),'bo-') ;
axis([0 N_trials m M]) ; grid on ;
set(findobj(gca,'Color','r'),'MarkerFaceColor','r') ;
set(gca,'xtick',(0:blocks_session:N_blocks)*trials_block) ;
xlabel('Trial') ; ylabel('ADJUSTED mean(median(RT))') ;
title(sprintf('%s, %d Ss; %s, %d Ss',...
      l_1st2nd{1},length(Ss_1st2nd{1}),l_1st2nd{2},length(Ss_1st2nd{2}))) ;
legend(l_1st2nd,'location','NorthEast') ;

Mean RT(Gr1+3,day1)=761, Mean RT(Gr2+4,day1)=807, adjustment=2*23 msec.

Contents

MLZ02_plot_dprime.m

Get started

Design parameters

Load the data

Quick-and-dirty plot of the individual-subject data

Median response times, individual data

Time- and condition-related variables -- See MLPilot02_params for details

Group dprime plot:

Plot group averaged d'

Plot group d' by easy / hard delta

Plot group d' for delta 11 (easy_dprime2)

Plot group d' for delta 8 (hard_dprime2)

Plot group d' for delta 11 (easy_dprime2)

Sbj Order Learning Effect Scatter plot

Group Order Learning Effect bar plot

Group Order Learning Effect bar plot

Main effect of Order on dprime:

Group d'prime error plot

Group criterion plot:

Group RT plot

Main effect of Order on RT

Adjusted group RT plot